From Artificial Intelligence Adoption to Business Performance: The Role of AI Capability, Dynamic Capabilities and Responsible AI Governance in Emerging-Market SMEs
Published: 2026-11-19
Artificial
intelligence (AI) is widely expected to improve firm performance, yet evidence
that adoption alone creates sustained value remains mixed. This article
develops an integrated model explaining how AI adoption translates into
performance among small and medium-sized enterprises (SMEs) in emerging
economies, and specifies an empirical design to test it. Drawing on the
Resource-Based View (RBV), Dynamic Capabilities Theory (DCT), and the
Technology–Organization–Environment (TOE) framework, the model positions AI
adoption as an antecedent of multidimensional AI capability, which shapes
performance through two parallel mechanisms: dynamic capabilities (sensing,
seizing, and reconfiguring) and human–AI collaboration. Responsible AI
governance is theorized as a boundary condition that strengthens the conversion
of AI capability into performance. Nine hypothesis sets are derived, alongside
a proposed cross-sectional survey of Ethiopian SMEs analysed via partial least
squares structural equation modelling (PLS-SEM) and supplemented by qualitative
interviews. The study shifts scholarly attention from adoption as a discrete
event to capability orchestration, integrates a micro-level collaboration
mechanism, and frames governance as a value-enabling condition rather than a
purely compliance-driven constraint. Because no empirical results are reported
yet, the model is presented as a structured blueprint for testing.